REDUCE

CONSERVE OPTIMIZE

EXPLORE

Department of Environmental Economics Pakistan Institute of Development Economics, Islamabad, Pakistan

Department of Environmental EconomicsWorking Paper No. 7

CORE

Impact of Dust Pollution on Worker’s Health in Textile Industry: A Case

Study of Faisalabad, Pakistan

Nazia MehwishUsman Mustafa

Department of Environmental Economics

Working Paper No. 7

Impact of Dust Pollution on Worker’s Health in Textile Industry: A Case

Study of Faisalabad, Pakistan

Nazia Mehwish Pakistan Institute of Development Economics, Islamabad

and

Usman Mustafa

Pakistan Institute of Development Economics, Islamabad

PAKISTAN INSTITUTE OF DEVELOPMENT ECONOMICS

ISLAMABAD

Note: This working paper is published in continuation of the Working Paper Series published earlier

by Centre for Environmental Economics and Climate Change (CEECC), Pakistan Institute of

Development Economics (PIDE), Islamabad.

All rights reserved. No part of this publication may be reproduced, stored in a retrieval system or

transmitted in any form or by any means electronic, mechanical, photocopying, recording or

otherwise without prior permission of the Publications Division, Pakistan Institute of Development

Economics, P. O. Box 1091, Islamabad 44000.

© Pakistan Institute of Development Economics, 2016.

Pakistan Institute of Development Economics

Islamabad, Pakistan

E-mail: publications@pide.org.pk

Website: http://www.pide.org.pk

Fax: +92-51-9248065

Designed, composed, and finished at the Publications Division, PIDE.

C O N T E N T S

Page

Abstract v

1. Introduction 1

2. Data Description and Methodology 3

2.1. Description of Study Area 3

2.2. Data Collection and Survey Design 3

2.3. Methodology 5

3. Results and Discussion 8

3.1. Results of Confirmatory Factor Analysis 8

3.2. Results of the Structural Regression Model 10

4. Conclusion and Policy Implications 16

4.1. Conclusion 16

4.2. Policy Implications 17

References 18

List of Tables

Table 2.1. Measurement Model (Confirmatory Factor Analysis) 5

Table 2.2. Structure Regression Model 8

Table 3.1. Results of the Measurement Model (Confirmatory Factor

Analysis) 11

Table 3.2. Results of Structural Regression Model 13

List of Figures

Figure 3.1. Confirmatory Factor Analysis of Respiratory Diseases 9

Figure 3.2. Path Diagram for Estimated Model Cost of Illness/

Structural Regression Model 12

ABSTRACT

The textile industry in Pakistan has largest share in the manufacturing

industry sector. The textile sector is one of the most polluting industrial sectors.

Cotton dust is present in the air during the handling and processing of cotton

which threatens the health of labours working in this industry. This study

estimated the impact of the dust pollution on workers’ health and cost of illness

in the textile industry of the Faisalabad, Pakistan. This was the cross sectional

research conducted among the 200 randomly selected textile workers. Both

health and opportunity cost of textile workers estimated by using the structural

equation model (SEM) with observed and latent variables. This paper used the

confirmatory factor analysis (CFA) to interlink the latent factor diseases with

their indicators symptoms. The results of the SEM model on age, respiratory

diseases, overtime work, duration of employment, use of masks and dust level

are significant. Study finds that 62 percent workers beard the cost of illness and

43 percent workers miss the work last two weeks. The 69 percent workers

reported that due to the illness their work performance in industry is not normal,

they suffered the problems during the work such as muscle aches, asthma,

cough; respiratory allergy etc. It also calculated the prevalence of the respiratory

diseases among the textile workers was high. Result revealed that 35.5 percent

workers had wheezing, 65.5 percent phlegm, 58 percent chest tightness, 72

percent had a throat irritation problems. Paper concluded that there is a high

prevalence of respiratory diseases among the textile workers due to the exposure

of dust pollution and if successfully reduction in dust pollution occurs in textile

industry; the worker will gained the benefits in term of the reduction in medical

cost and gain in terms of wages.

Keywords: Dust Pollution, Health Cost, Opportunity Cost, Respiratory

Diseases and Symptoms, Structural Equation Model (SEM),

Confirmatory Factor Analysis (CFA), Faisalabad, Pakistan

1. INTRODUCTION*

In the world the textile is the second largest industry after the agriculture

[Sangeetha, et al. (2013)]. The textile industry in Pakistan holds the leading role

in the manufacturing industry sector. This sector comprises almost 8.5 percent to

the GDP, provides employment to more than 46 percent of the manufacturing

sector labour force and contribute major share in foreign exchange earnings for

the country. Further, Pakistan is the 8th largest exporter of textile products in

Asia, 4th largest producer of cotton, with the third largest spinning capacity in

Asia after China and India, and contributes 5 percent to the global spinning

capacity. This textile sector has an overwhelming impact on the growth and

development of the Pakistan’s economy [Government of Pakistan (2015)].

The textile sector is one of the most polluting industrial sectors [Banuri

(1998)]. The textile industry is also associated with a number of environmental

problems such as water pollution, soil pollution, noise pollution and air/ dust

pollution. Among the different textile pollution, cotton dust pollution is the most

important in terms of health effects [Camici, et al. (1978); Beck, et al. (1983)].

Cotton dust pollution, lint, and particulates impact the working environment and

affect the workers respiratory system. The dust level or quantity small particles

of cotton lint infiltrate in to the respiratory system of the human beings and

cause the respiratory diseases and symptoms among the workers. These

respiratory diseases cause some limitations such as increase the doctor visit, loss

in workdays and reduction in quality of work. Cotton mills are polluted with the

different type of the cotton dust hazards. Workers which work in these mills are

associated with the number of the respiratory diseases and symptoms e.g. chest

tightness, shortness of breath, increased cough and phlegm, wheezing, asthma,

tuberculosis, lung-function loss, eye sight problems and skin diseases

[Salvaggio, et al. (1986)].

Due to such kind of the illnesses, the number of the problem arises for

workers, such as the loss in income, the loss in the good health and suffering

pain due to the illnesses, increase in the health care cost, poor work performance

due to the bad health, medicine and doctor fees, loss in workers’ productivity

due to the cost of illness, increase the social conflicts in the workers families,

reduction in the production, and negative impact occur on moral of the other

workers. World Health Organisation showed that eight hundred thousand labour

Acknowledgement: Authors express deepest thankfulness to South Asian Network for

Development and Environmental Economics (SANDEE) for assistant in providing the data on the

workers respiratory diseases, socio-eco factors, health and medical status of the textile workers.

2

forces in the cotton and in rope making industries are suffering due to the dust

pollution so in such situation there is the high probability of prevalence of the

respiratory and byssinosis diseases [WHO (2008)]. Byssinosis is the name of the

brown lung disease which is caused by the cotton dust.

Mostly, workforce in the developing countries is uneducated and most of

them are involve in the cotton handling process in the textile industry. So, that

there is the high probability of prevalence of the respiratory diseases like

asthma, cough, fever and brown lung diseases. The intensity of these diseases

among these workers is high due to their illiteracy and they are not aware of the

use of the safety gadgets such as wearing the face masks, use of the respirator

etc.

The long term exposure to dust pollution can cause the respiratory

diseases [WHO (2008)]. Cotton dust exposure at textile industries adds

considerably to the occupational burden of disease globally [Nafees, et al.

(2013)]. The relationship between the cotton dust in the textile industries and the

prevalence of the respiratory illnesses among the industrial workers has been

well documented [Memon, et al. (2008); Farooque, et al. (2008); Anjum, et al.

(2009); Nafees, et al. (2013)]. The objective of the study wasto provide the

strong evidence of Byssinosis and other respiratory diseases in textile industry

and to estimate the workers’ cost of illness associated with dust pollution among

textile workers in Faisalabad, Pakistan.

There are number of respiratory diseases and difficult to identify with

clinical tests, therefore, the study identified the respiratory diseases through their

symptoms. These latent respiratory diseases are linked with their indicators

symptoms because which are verified by chest specialist. The chest specialist

provided consultancy/expert views in identifying the related diseases that may

be caused by the dust pollution in the textile mills. This is a social scientist study

and adopting the medical test and professional is very expensive and time

consuming, therefore, for the identification of the diseases study used the

confirmatory factor analysis model.

So in this context, this study estimated the cost of illness. For the

estimation of the cost of illness study use the structural equation models (SEM)

with observed and latent variables. The estimation of the cost of illness by using

the structural equation models (SEM) is different from other studies because

costs are directly linked to the respiratory diseases but diseases are not

predictable among the humans but its reflected with their symptoms (indicators),

so for solving this causal problem we used the confirmatory factor analysis

(CFA) model to predict the different latent factors diseases with their different

symptoms indicators and then we estimate the impact of the respiratory diseases

and different demographic variables on the endogenous variables health and

opportunity cost.

3

Paper is organised as follow, Section 2 provides data description and

methodology, Section 3 results and discussion and Section 4 is about conclusion

and policy implication.

2. DATA DESCRIPTION AND METHODOLGY

2.1. Description of Study Area

Faisalabad is the big industrial town of Pakistan. Faisalabad is known as

the “city of textile” because the numbers of the textile industry are clustered in

the district Faisalabad. Textile industry is the backbone of the Faisalabad

economy. Sitara Textile Mills and Lyallpur Cotton Textile Mills, Kohinoor

Textile Mills, Crescent Textile Mills, Rahmania Textile Mills, Masood Textile

Mills, Aasim Textile Mills and Nishat Textile Mills are the famous textile

industries in the district Faisalabad. These industries have the major role in the

development of the Faisalabad economy and 5 billion dollars are the annual

textile exports arrive from the Faisalabad textile industries [Khan (2013)]. There

are 612 large industrial units out of which 248 are textile units [Malik (2010)].

Due to the rapid development in the textile sector, Faisalabad is known the

“Manchester of the Pakistan” [Khan (2013)]. At the same time, city has high

energy utilisation and dense transport system which cause dust pollution. Dust

pollution can lead to a variety of health as well as environmental problems. One

of the main sources of dust pollution is the textile industry.

2.2. Data Collection and Survey Design

The objective of the paper is measuring the impact of dust pollution on

workers’ health and related cost of illness, so for this purpose we needed to

collect the primary data. The primary data collected through questionnaire and

face to face interview with workers. The questionnaire was developed on the

basis of the questionnaires of the health cost and respiratory symptoms studies

[Gupta (2006); Murty, et al. (2003); Adhikari (2012); Chowdhury and Imran

(2010); Bogahwatte and Herath (2008); Memon, et al. (2008); Farooque, et al.

(2008); Nafees, et al. (2013); Hinson, et al. (2014)]. The questionnaire has two

parts, the first part is the general workers survey questionnaire (socio-economic

characteristics of workers) and second part is Health Diary Questionnaire

[Kamat and Doshi (1987); Gupta (2006); Chowdhury and Imran (2010);

Bogahwatte and Herath (2008)].

The general worker survey questionnaire includes information on work

characteristics and socio-economic factors. Variable in socioeconomic

characteristics are age, education, monthly income, gender and smoking history

of workers in textile industry and variables in work characteristics are duration

of employment, work section, mask use and working hours. A notable feature

of this study is that it uses data from health diaries.

4

The second part of the questionnaire is health diary questionnaire. The

health diary questionnaire includes the detail information on respiratory illnesses

and symptoms of textile workers. Through Health Diaries we collected the data

on respiratory diseases and symptoms, health cost, dust level, opportunity cost

and data on workday lost in the last two weeks. Gupta (2006); Adhikari (2012);

Chowdhury and Imran (2010); Bogahwatte and Herath (2008) used the similar

health diary technique to collect the data on health and medical status of the

respondents and associated cost.

Health cost data consist the expenditure on medicines, hospital stay,

pathological tests, doctor fees and accompanying person cost. The opportunity cost

includes the cost of the workday miss in the period of the last two weeks. The study

used the multistage sampling technique to select workers. This is the cross sectional

research conducted in the 3 spinning textile mills and randomly selected the 200

workers. These three spinning textile mills associated with the All Pakistan Textile

Mills Association (APTMA) Faisalabad. The previous literature has shown that level

of dust is much higher in the spinning section of the textile industry. “Spinning is a

manufacturing process where fibres are converted into yarn”. Spinning is the

twisting together of drawn out strands of fibres to form yarn.

The data was collected through the permission of the industry

administration. The actual data on dust pollution is not collected, since it is not

useful for the industry. The measurement of dust pollution in each factory and in

each section of spinning mill is very costly and monotonous job. Further they do

not allow outsider to do this kind of activity as it is against the industry. So here

the data on variable dust level is available through the self-reported by the

textile workers. Dust level is measured on a five point likert scale from high dust

pollution to low dust pollution.

Study used the health diary data based on the “upcoming two weeks”.

Same two weeks health diary data used the Bogahwatte and Herath (2008) in

their study but their health diary data based on the “two week recall”.

This study collected the data in two phases:

In the first phase, the data collected on the “upcoming two weeks”

based health diary questionnaire.

The second phase was the collection of the information on general

worker survey.

The data on health diary questionnaire has collected through the

following procedure. The first health diary was translated in to Urdu language.

Then the health diary questionnaires were dropped to the workers and requested

them to fill the Urdu health diary questionnaire for the upcoming fifteen days. In

health diary questionnaire workers mentioned the information on respiratory

diseases, cost of illness and workday lost. But the general worker survey data

was collected through face to face interview.

5

2.3. Methodology

To estimate the cost of illness we used the Structural Equation Model

(SEM).

2.3.1. Structural Equation Modelling (SEM) with Observed and Latent

Variables

To test the cost of illness of the textile workers we employ structural

equation model (SEM) with observed and latent variables. SEM is a “statistical

technique for testing and estimating causal relationships amongst the variables,

some of which may be latent using a combination of statistical data and

qualitative causal assumptions. Latent variables (also known as unobserved

variables, hypothetical variables or hypothetical constructs) are variables that are

not directly observed but are reflected from other variables that are observed and

directly measurable.”

SEM deal with the latent and observed variable. The observed or

measured variables are those variables which is directly measurable. Latent

variables are not directly measured able but it is inferred from the observed

variables and confirmatory factor analysis (CFA) deal the latent variables.

The structural equation model (SEM) basically consist two parts, one part

is the measurement model (confirmatory factor analysis) which shows the

relationship between the latent variables with their component indicators and

second part is structural model which assign the causal relationships among the

latent variables [Toma, et al. (2009)]. So for this purpose this study used the

structural equation model (SEM) in two stages,

Stage 1: is measurement model or confirmatory factor analysis (CFA).

Stage 2: is structural model or structural regression model (SRM).

Stage 1

2.3.1.1. Confirmatory Factor Analysis / Measurement Model

Diana (2000) stated that Confirmatory factor analysis (CFA) is “a

statistical technique used to verify the factor structure of a set of observed

variables. CFA allows the researcher to test the hypothesis that a relationship

between observed variables and their underlying latent constructs exists”.

In Confirmatory factor analysis (CFA) we identify the number of the

different factors which fundamentals in the data set and measured those

variables which is related to the latent variable. Mueller and Hancock (2007)

evaluate that the variation in dependent variables typically is not completely

explicable by the amount of variation in their particular causes or independent

variable. Confirmatory factor analysis (CFA) of the model is the part which

links each of the indicator variables with designated latent variable. In CFA,

indicators are specified to load on a single factor and consistent factor loading

6

are estimated correlations among the factor and its indicators and indicators are

generally correlated with each other but errors terms of the different indicators

are uncorrelated. In CFA per factor is fixed at least on one loading. Each set of

indicators are loaded onto a separate factor is given in the Figure 3.1.

The CFA is defined by the following system of the equations in Table 2.1

Table 2.1

Measurement Model (Confirmatory Factor Analysis)

Endogenous Variables Structural Equation Exogenous Variables

SBDW (X11) 111 11 EASTX X Asthma

SBODO (X12) 212 12 EASTX X

SBOT (X13) 313 13 EASTX X

SBOA (X14) 414 14 EASTX

Weaz (15) 515 15 EASTX

CT (16) 616 16 EASTX

LC (X21) 721 21 EBYSX Byssinosis

Fever (X22) 822 22 EBYSX

Swet (X23) 923 23 EBYSX

MA (X24) 1024 24 EBYSX

Phl (X25) 1125 25 EBYSX

Head (X26) 1226 26 EBYSX

EI (X31) 1331 31 ERAX Respiratory Allergy

TI (X32) 1432 32 ERAX

NI (X33) 1533 33 ERAX

CFT (X41) 1641 41 ECougX Chronic Cough

CMT (X42) 1742 42 ECougX

Table 2.1 implies a set of 17 structural equations from the measurement

portion of the model or confirmatory factor analysis (CFA). Here the X11=

shortness of breath during work, X12= shortness of breath one day off, X13=

shortness of breath other time. X14= shortness of breath always, X15=

wheezing, X16= chest tightness, X21= lung cancer, X22= fever, X23=sweating,

X24= muscle aches, X25= phlegm, X26= headache, X31= eye irritation, X32=

throat irritation, X41= cough few times and X42= cough many times. The

measured and latent variables in the current models are labelled X11 through

X42 and AST through Coug respectively.

7

In CFA system of model Asthma is the unobserved exogenous latent

variable which we cannot measure directly so we estimate it with the different

symptoms as X11, X12, X13, X14, X15 and X16. So through this system of the

CFA we estimated the one respiratory disease Asthma and similarly we

estimated the other unobserved latent respiratory diseases like a byssinosis,

respiratory allergy and chronic cough.

Stage 2

2.3.1.2. Structural Regression Model (SRM)

The structural model or structural regression model screening the

prospective causal dependence among endogenous and exogenous variables.

In this study we measured the cost of illness (OC and HC) through the help

of the SRM model because cost is directly associated with the dust pollution

illnesses but these diseases are unidentified. The actual data on these

diseases are not available from the textile workers. After consultation of the

Doctors and according to their advice then study interlink the respective

diseases with symptoms. Diseases are latent factor so that these diseases are

reflected by their symptoms. Medical studies show that respiratory illnesses

especially byssinosis are not directly observable but we reflect these

diseases with their cause (symptoms) and textile workers are not aware

about these diseases so we find these diseases among workers through their

symptoms. Memon, et al. (2008) and Manthur, et al. (2005) observed that

byssinosis is such respiratory disease which occur among the textile workers

in minimum of the 10 years of the work in the textile industry and its

symptoms among the workers occur within the 3-5 years duration of the

work in textile industry.

So for solving this causal problem first we predicted the different latent

diseases with their different symptoms indicators and then estimated the impact

of the respiratory diseases and different demographic variables on the

endogenous variables health and opportunity cost.

The SRM is defined by the following system of the equations in Table

2.2.

The structural regression model consists of two endogenous variables,

opportunity cost (Cost of workday miss) and health cost (medication cost)

and the number of the exogenous variable. We estimated the health and

opportunity cost with the help of the number of latent and other variables of

the model. Table 2.1 and 2.2 lists the all 19 structural equations and

connected dependent and independent variable that evaluate the path model

in Figure 3.2.

8

Table 2.2

Structure Regression Model

Endogenous

Variable Structure Equation Exogenous Variable

Health cost

Opport-

unity cost

2.....................

1.....................

1211109

87654321

12119

87654321

10

SmOTUMDW

DLEduIncAgeCougRABYSASTOC

SmOTUMDW

DLEduIncAgeCougRABYSASTHC

Asthma, Chornic

Cough,

Byssinosis,

Respiratory Allergy

Age, income,

education, overtime,

use of mask, dust

level, duration of

work, smoking

3. RESULTS AND DISCUSSION

This section of the study describes the results obtained from structural

equation modelling, which is generally categorised as two stages modelling

first stage is CFA and second stage is SRM. As in this study we have the

different number of the unobserved latent variables so at first stage we apply

the CFA as a measurement model approach to specify the reflection of

asthma, byssinosis, respiratory allergy, and chronic cough in terms of their

relevant indicators (symptoms). Once the CFA has been estimated then we

in the second stage SRM is applied to shows the effect of the latent factors

and different socio-economic variables on endogenous variables health and

opportunity cost.

3.1. Results of Confirmatory Factor Analysis

Statistical techniques such as factor analysis, explanatory or

confirmatory have been widely used to examine the number of latent

constructs underlying the observed respondents and to evaluate the adequacy

if individual item or variables as indicator for the latent constructs they are

supported to measure. In our case we have four factors, asthma, byssinosis,

respiratory allergy and chronic cough. Each factor is reflected with list of

indicators.

The indicators are generally correlated with each other but errors terms of

the different indicators are uncorrelated. In CFA per factor is fixed at least on

one loading. Each set of variables loaded onto a separate indicator is presented

in the Figure 3.1.

Medical literature on respiratory illnesses reported that respiratory

diseases are not directly observed this study observed it by the different

symptoms through the confirmatory factor analysis factor loading.

9

Fig. 3.1. Confirmatory Factor Analysis of Respiratory Diseases

In Figure 3.1 the measured variables are point out by in rectangular form

(X11to X42) and latent variables are point out by ellipses (BYS, AST, COUG,

and RA). There are four endogenous latent variable as AST (asthma), BYS

(byssinosis), RA (respiratory allergy) and Coug (chronic cough) and we loaded

these variables with their symptoms because these diseases are not directly

observable its depend on different factors or indictors. As AST (asthma) is

indicated with the following set of indicators X11, X12, X13, X15, X16, X17

and X24. In this Figure, the one headed arrows show the direct causal effect

10

hypothesised from one variable effect to another variable. In Figure 3.1, the

residual terms (e) point out that reflection to indicators is influenced by the

causes other than the latent variables.

There is nonzero covariance may exist between e1 to e5. The analysis

of covariance is based on the implicit assumption that indicators are

measured as deviations from their means (i.e., all indicator means equal

zero). It seems that CFA model justifiable the causes with latent factors and

it allow the residual terms of resultant respiratory symptoms to freely

covary. Covariance exit between the two headed arrows among the two

variables. In SEM, the latent variables are the free of error term. Factor

loadings estimate the direct effect of factors on indicators. This factor

loading shows that we estimate the respiratory disease asthma through the

following symptoms. This factor loading is doing for this purpose to access

the impact of this factor loading on the opportunity and health cost because

diseases are not directly identified and also not statistically significant on

health and opportunity cost. The study also aims to know about the health

status of the textile workers.

The Table 3.1 shows the results of number of the respiratory illness

and symptoms. The results indicate that the relationships between the latent

and observed variables diseases and symptoms have the significant positive

relationship. In SEM model the variables are supposed mean centred, so

eradicating the requirement for the intercept term in the mean centred data,

the indictors X16, X26, X33 and X42 are used as the indication variable for

their respective variable loading and their factor or variable loading are not

freely statistical measured but we fixed to 1. It is sufficient condition for the

identification of the measurement model and AMOS software automatically

fixed the one indicator. Asthma, byssinosis, respiratory allergy, and chronic

coughing are statistical significant at 5 percent level.

This shows that the textile workers are suffering these diseases due to the

exposure of the dust pollution. This setup is called confirmatory factor analysis.

This explains that the relationships between these variables are driven by

underlying latent variables.

The results of the Figure 3.1 and in the Table 3.1 shows that standardised

factor loading was statistically significant, it similarly shows that the

correlations between error terms and among latent variables these factor

loadings were also statistically significant.

3.2. Results of the Structural Regression Model

Structural regression model is the second stage of the structural equation

model (SEM). In structural regression model (SRM) we estimated the SRM with

the measurement model proven in first stage but with the different other

exogenous variable of the SRM.

11

Table 3.1

Results of the Measurement Model (Confirmatory Factor Analysis)

Indicators

Estimates

AST

Estimates

BYS

Estimates

RA

Estimates

Coug

X16 <--- AST 1.000

X15 <--- AST 0.924*

(.163)

X14 <--- AST 1.207*

(.197)

X13 <--- AST 0.309*

(.125)

X12 <--- AST 0.531*

(.127)

X11 <--- AST 0.317* (.119)

X24 <--- AST 0.251*

(.129)

X26 <--- BYS

1.000

X25 <--- BYS

2.053*

(.388)

X24 <--- BYS

1.306* (.290)

X23 <--- BYS

1.674*

(.416)

X22 <--- BYS

0.526*

(.197)

X21 <--- BYS

2.012*

(.382)

X12 <--- BYS

1.422* (.295)

X11 <--- BYS

1.208*

(.265)

X13 <--- BYS

1.362*

(.291)

X15 <--- BYS

0.886*

(.214)

X33 <--- RA

1.000

X32 <--- RA 0.910*

(.083)

X31 <--- RA

0.952*

(.211)

X22 <--- RA

0.634*

(.164)

X26 <--- RA

0.486*

(.147)

X42 <--- Coug

1.000

X41 <--- Coug

0.918*

(.217)

X16 <--- Coug

0.441*

(.109)

CMIN=470 d.f = 111 CMIN/DF = 4.236 CFI = 0.691 , SE in parenthesis.

12

So we estimated the structural Equations (1) and (2) in Table 2.2 through

the structural regression analysis. The diagrammatic representation of the

structural model is given in Figure 3.2 and estimates are presented in Table 3.2

In below in the Figure 3.2 which presents the path diagram for the

estimated model on the health and opportunity cost. This path diagram includes

the set of 17 structural equations from the confirmatory factor analysis (Table

2.1) and two equations from the structural model or SRM (Table 2.2).

Fig. 3.2. Path Diagram for Estimated Model Cost of Illness /

Structural Regression Model

13

In Figure 3.2 path diagram include the two endogenous variable as health

and opportunity cost and twelve exogenous variable, namely asthma (AST),

byssinosis (BYS), respiratory allergy (RA), cough, age, income (Inc), education

(Edu), duration of work (DW), smoking (Sm), use mask (UM), dust level (DL)

and overtime work (OT).

Table 3.2

Results of Structural Regression Model

Health Cost Opportunity Cost

AST 742.74

(0.002)

490.5

(0.000)

BYS 250.32

(0.008)

485.49

(0.137)

RA 798.18

(0.000)

579.15

(0.000)

Coug 407.14

(0.000)

164.93

(0.000)

Sm 178.74

(0.000)

28.01

(0.249)

Age 2.08

(0.074)

7.14

(0.000)

Inc 0.012

(0.098)

0.010

(0.002)

Edu 5.14

(0.481)

0.019

(0.995)

DW 9.87

(0.044)

6.43

(0.003)

OT 6.895

(0.041)

6.03

(0.000)

UM 0.626

(0.09)

46.55

(0.05)

DL 105.4

(0.000)

4.3

(0.758)

CMIN =1864, d.f1 =382, CMIN/DF=4.8, CFI=0.635 P-values in parenthesis.

In Table (3.2) shows the results of structural regression model (SRM)

which measured the impact of latent diseases and other variables on health and

opportunity cost. The latent variable chronic diseases, Byssinosis, chronic

1d.f= P(P+1)/2-K.

P= number of items, K= number of parameters.

14

cough, asthma and respiratory allergy have positive coefficients are significant

at one percent level, meaning thereby that workers with these conditions have

higher medical and opportunity costs.

One respiratory disease byssinosis are statistical significant to health cost

(HC) but not significant to opportunity cost (OC) with expected positive sign

(P=0.137). However byssinosis is significant with the OC at the univariate level

but it is insignificant at the multivariate level when we attached the other

variables in the structural equation model (SEM). Byssinosis is brown lung

disease which occurs due to the dust pollution exposure [Hinson (2014)]. As

workday lost is one censored variable so the present study results shows that

may worker not miss the work last two weeks due to the byssinosis and so not

bear the opportunity cost. Study has the some policy and time restrictions so

collected the data only for last two weeks, thus the impact of this limitation

occurs on results.

Age is significant at one percent level (P = 0.000) to OC but at ten percent

to HC with the expected positive sign and age coefficient has the positive sign

which shows that old age workers have the more medication and opportunity cost

because they have work in the same industry from the number of the years as

survey data shows that some workers have the 20–25 years work experience and

also these old age workers are mostly uneducated. The uneducated workers are not

follows the safety health measures so these workers suffer the number of the

respiratory diseases and bear the cost of illness. It is well established fact that age

has dominant effect on the health and working capacity of a man. A young person

can work for more time and not much susceptible to diseases than old one. Old

people are highly affected from respiratory diseases. So, working capacity of

different age group workers is different. Age group of respondents is given an

important position in the present study.

The coefficient income has the positive sign which is statistical significant

at five percent (P=0.002) and ten percent level (P=0.09) in both opportunity and

health cost respectively. The positive sign of the coefficient income mean higher

income and higher income or positive income associated with the higher medical

and economic expenditure. Adhikari (2012) reported the same results about

positive income effect on mitigation cost. Literatures highlight that which workers

have the higher income they spend more income on their health. Same results

proposed by the Chowdury and Imran (2010) that better standard of living and

positive income have positive effect on the medical expenditure.

The variable smoking has a significant positive relationship on health cost

of the textile workers. Smoking is injurious to health is a very well-known

proverb. Smoking is has very bad effect on the respiratory system of a human

body. It causes various diseases like asthma, shortness of breath, cough, T.B,

and lungs carcinoma. Ghasemkhani, et al. (2006) also described that smokers

have the higher prevalence of the respiratory diseases and symptoms.

15

So smoking has the significant impact on the health cost of the workers

which mean smoker workers tolerate more medical expenses as compared to the

non-smoker workers. As a result, smoking affects the working capacity of

workers thus they bear the additional cost. The results of the SEM indicate that

smoking is not significant to opportunity cost since opportunity cost is workday

miss cost may workers not miss their work last two weeks due to the effect of

the habit of the smoking. A same result was declared to the Gupta (2006) in his

study he did not found the significant relationship between cost and smoking

habit.

The duration of employment in the textile industry also significant at the

5 percent level on both opportunity and health cost. The prolonged period of

employment in the dusty section of the textile industry have the negative impact

on the worker health [Vays (2000)].

Ghasemkhani, et al. (2006) and Memon, et al. (2008) indicate that long

duration of employment in the textile industry have the major role in the

development of the respiratory diseases among the textile workers.

Ghasemkhani, et al. (2006) described that the incidence of the respiratory

diseases increased with the age and duration of employment. Farooque, et al.

(2008) observed that those workers had the more respiratory illnesses which

work in the same industry last 10 to15 years. The present study pragmatic that

around the 62 percent workers had the duration of employment in the textile

industry between 6 to 15 years. Therefore the long period of the employment in

the textile industry has the impact on the worker health and thus worker bear the

more health and opportunity cost.

The relation of use mask with opportunity and health cost is significant at

5 percent (P=0.05) and at 10 percent level (P=0.09) respectively but it

coefficient positive sign is surprised for the study. Use of mask is very good

habit during work in textile industries. It helps to overcome the allergy and other

respiratory illness in workers but the results of the survey data shows that 77

percent workers not used the mask during the work.

The literature and general evidence shows that there is positive

association between dust level and prevalence of the respiratory diseases. Dust

level has the positive relation with the health and opportunity cost. It is

statistical significant at 1 percent with the HC (P=0.000) but not significant with

OC (P=0.758). The positive sign of the coefficient shows that higher the dusts

level higher the health cost of the textile workers. Su (2003) highlighted that

high prevalence of the byssinosis among workers significantly associated with

the high dust level, 45 percent workers had the symptoms of byssinosis due to

high dust exposure. The above Table result shows that dust level is not statistical

significant to opportunity cost but it has positive sign (P = 0.758). The positive

sign of the coefficient of dust level indicate that higher the level of dust is

positively associated with the opportunity cost (high level of dust increase the

cost).

16

Some our demographic variables like gender, education and heat or

temperature are not significant to the cost of illness. Gender and temperature are

also not significant in Gupta (2006) study. In present study, the frequency of

female workers in textile industry is very less as compared to the male workers,

so for this reason the variable gender is insignificant.

Structural regression models are usually evaluated in term of how well

they fit the data. The goodness of fit of the models measured by the usually chi

square statistics.

In order to check to check the overall goodness of fit in SEM model

usually we are using the two criteria of chi-squared statistic one is the

CMIN/DF2 and second is the comparative fit index (CFI).In general CFI ranges

from 0 to 1and the larger value of CFI is indicating good model fit, the

acceptable range of the CFI is between 0 to 1 and it is good fit model if CFI is

near to 0.9 or 1. In our study CFI =0.63 is close to 0.9 or 1 which indicated that

model good fit the data and CMIN/DF=4.8 which is the less than to the table

value 5, the table value roughly is closed to 5 thus H03 is acceptable so the

calculated value should be less than the table value. When considered the

following indices then data model fit is considered acceptable.

4. CONCLUSION AND POLICY IMPLICATIONS

4.1. Conclusion

This study is conducted in the textile industry of the Faisalabad. The

primary goal of the study was to assess the impact of dust pollution on worker

health, productivity and cost. This study found the high prevalence of the

respiratory diseases and symptoms among the textile workers. This shows that

workers lose their productivity and working capacity in the textile industry. The

69 percent workers reported that due to the illness their work performance in

industry not normal, they suffered the problems during the work such as muscle

aches, wheezing, chronic cough, asthma, respiratory allergy etc. This indication

that workers are exposed to the dust pollution illnesses. Study also measured the

prevalence of the respiratory diseases among the textile workers. Study revealed

that (35.5 percent) workers had wheezing, (65.5 percent) phlegm, (58 percent)

chest tightness, (72 percent) had a throat irritation problems.

Study observed that almost 30 percent workers suffered the health cost

(HC) between the 1 to 500 rupees, 15 percent in the range of 600 to 1000 rupees,

15 percent among Rs 1100 to 2000 and 2.5 percent between 2100 to 2500 rupees

in the last two weeks. Due to the last two weeks workday lost 41 percent

workers beard the opportunity cost in the range of Rs 500 to 1500.

2CMIN : Chi-Square minimum, CFI: Comparative fit index, d.f: degree of freedom. 3In SEM , H0: model fit the data.

H1: model does not fit the data.

17

The research paper provided the conclusive evidence on the opportunity

and health cost. The reduction in dust pollution positively impact on the health

and opportunity cost status of the workers. The results evidently show that

reduced in dust will reduced the respiratory diseases and due to the reduction in

respiratory diseases will reduced the health and opportunity cost.

4.2. Policy Implications

The safe workplace environment gives the indication of the good health,

increase in productivity of the workers and also enhances the production of the

industry. The Govt. should to take such inventiveness measure which reduced

the dust from the textile industry and promote the safe and clean environment

and also take such initiative to measure the dust level within the industry on

overtime.

Govt. should to give the subsidy in polluted industries such as textile;

construction industries etc. for the installation of the modern technology and

make the law and policies for such type of industries to check the level of dust in

these industries on the daily or weekly basis. If clean working environment

promote in our industries we observed the great change in the health status of

the worker and change in economic condition of the worker.

Factory owners must needed to adopt such technologies in industry which

automatically abate the dust pollution during the work. Factory owner also

should to compensate the affected workers for their health losses in terms of the

reduction in the opportunity cost which they bear due to absent from work due

to the dust pollution sickness. Industry administration has to arrange the

programmes and training workshop about awareness of the dust pollution

hazards among the workers.

The textile industry worker should to follow the preventive safety

measures; they must wear the mask during the work and adopt such useful

measures which prevent them from the dust during the work.

The industry proprietors have the great indication on the business as usual

basis to potentially increase their savings in terms of the provision of the healthy

work environment. When workers have the less prevalence of the respiratory

diseases, they will have the more working capacity and they will increase the

production level of the industry. Industry owner should to increase their

economic performance in terms of the safe provision of the working

environment and this effort of the firms mean they play a vital role in the

sustainable development of the country. The evidence shows that safe

environment have the significant positive impact on the workers health as well

as on their social life. The healthy and mentally relax workforce play the vital

role in the sustainable development of the country.

The purpose of the present study is to aware the policy makers, Govt and

other stakeholders about the tangible opportunity and health costs that workers

18

bear due to the dust pollution in the textile industry so the policy-makers should

to adopt such interventions which tackle the dust pollution from the textile

industries of the Pakistan.

REFERENCES

Adhikari, N. (2012) Measuring the Health Benefits from Reducing Air Pollution

in Kathmandu Valley: Kathmandu: South Asian Network for Development

and Environmental Economics (SANDEE).

Ahmed Nafees, Masood Fatmi, and N. Sathiakumar (2013) Pattern and

Predictors for Respiratory Illnesses and Symptoms and Lung Function

among Textile Workers in Karachi, Pakistan. Occupational and

Environmental Medicine 70:2, 101–109.

Alemu, K., A. Kumie, and G. Davey (2010) Byssinosis and other Respiratory

Symptoms among Factory Workers in Akaki Textile Factory, Ethiopia.

Ethiopian Journal of Health Development 24:2.

Anjum, Mann, and Anjum (2009) Health Concerns among Workers in Weaving

Industry: A Case Study of Tehsil Faialabad, Pakistan. J. Agric. Soc. Sci. 5:

106, 8.

APTMA, All Pakistan Textile Mills Association.www.aptma.org

Ayres, J. G., R. Boyd, H. Cowie, and J. F. Hurley (2010) Costs of Occupational

Asthma in the UK. Thorax, thx. 2010.136762.

Beck, G. J., Schachter, L. R. Maunder, and R. S. Schilling (1982) A Prospective

Study of Chronic Lung Disease in Cotton Textile Workers. Annals of

Internal Medicine 97:5, 645–651.

Bogahawatte, C. and J. Herath (2008) Air Quality and Cement Production:

Examining the Implications of Point Source Pollution in Sri Lanka.

Byrne, B. M. (2013) Structural Equation Modeling with AMOS: Basic

Concepts, Applications, and Programming: Routledge.

Dasgupta, P. (2004) Valuing Health Damages from Water Pollution in Urban

Delhi, India: A Health Production Function Approach. Environment and

Development Economics 9:01, 83–106.

Farooque, M. I., B. Khan, F. Aziz, M. Moosa, M. Raheel, S. Kumar, and F. A.

Mansuri (2008) Byssinosis: As Seen in Cotton Spinning Mill Workers of

Karachi. Journal-Pakistan Medical Association 58:2, 95.

Ghasemkhani, M., M. Kumashiro, M. Rezaei, A. R. Anvari, A. Mazloumi, and

Sadeghipour (2006) Prevalence of Respiratory Symptoms among Workers in

Industries of South Tehran, Iran. Industrial Health 44:2, 218–224.

Gupta, U. (2008) Valuation of Urban Air Pollution: A Case Study of Kanpur

City in India. Environmental and Resource Economics 41:3, 315–326.

Hinson, A., V. Schlünssen, G. Agodokpessi, T. Sigsgaards, and B. Fayomi

(2014) The Prevalence of Byssinosis among Cotton Workers in the North of

19

Benin. The International Journal of Occupational and Environmental

Medicine 5: (4 October), PII 448, pp. 194–200.

Malik, N. (2010) Perspective of Occupational Health and Safety in Textile

Industry. University of Agriculture, Faisalabad.

Mathers, C., D. M. Fat, and J. Boerma (2008) The Global Burden of Disease:

2004 Update: World Health Organisation.

Memon, I., A. Panhwar, D. K. Rohra, S. I. Azam, and N. Khan (2008)

Prevalence of Byssinosis in Spinning and Textile Workers of Karachi,

Pakistan. Archives of Environmental and Occupational Health 63:3, 137–

142.

Mishra, A., S. Rotti, A. Sahai, and K. Narayan (2003) Byssinosis among Male

Textile Workers in Pondicherry: A Case-control Study. National Medical

Journal of India 16:2, 70–72.

Punjab, Govt. of (2014) Development Statistics of Punjab. Bureau of Statistics,

Planning and Development Department, Lahore.

Sangeetha, B., M. Rajeswari, S. Atharsha, K. S. Sri, and S. Ramya (n.d.) Cotton

Dust Level in Textile Industries and Its Impact on Human. International

Journal of Scientific and Research Publications 780.

Saoji, A. V., D. Aniruddha, K. Meenal, N. Jaydeep, and A. B. Mudey (2010) To

Study the Prevalence of Chronic Respiratory Morbidities and Related

Epidemiological Factors among Spinning Mill Workers. Global Journal of

Health Science 2:2, 111.

Su, Y.-M., J.-R. Su, J.-Y. Sheu, C.-H. Loh, and S.-H. Liou (2003) Additive

Effect of Smoking and Cotton Dust Exposure on Respiratory Symptoms and

Pulmonary Function of Cotton Textile Workers. Industrial Health 41:2, 109–

115.

Suhr, D. D. (2006) Exploratory or Confirmatory Factor Analysis? SAS Institute

Cary.

Toma, L., A. McVittie, C. Hubbard, and A. W. Stott (2011) A Structural

Equation Model of the Factors Influencing British Consumers’ Behaviour

Toward Animal Welfare. Journal of Food Products Marketing 17:2-3, 261–

278.

World Health Organization (1986) Early Detection of Occupational Diseases.